Internships

Algorithm Intern (Robotics Direction)

Beijing

Internship

Robotics Algorithm Direction

Students in relevant majors

Job Description

We hope you will participate in algorithm research, experimental design, and model iteration in the field of robotics, currently focusing on humanoid robot-related issues. This position is suitable for students who have a clear interest in algorithms, mathematics, robot learning, and scientific research exploration. You will be involved in the entire process from literature understanding, solution implementation to experimental verification, and collaborate with system colleagues to promote the application of algorithms on real platforms.

Participate in the research, implementation, experimentation, and optimization of algorithms in the field of robotics, advancing verifiable research results around specific problems.
Participate in literature research, solution reproduction, experimental design, result analysis, and iterative improvement in scientific research projects.
Participate in robotics data processing, model training, inference verification, and experimental evaluation, promoting the transformation of algorithms from concepts to executable results.
Collaborate with system colleagues to advance the deployment, testing, and integration of algorithms on real robot platforms, focusing on the combination of algorithm performance and system constraints.
Engage in research and exploration in the field of humanoid robots, supporting the continuous improvement of models, data, experimental processes, and evaluation methods.

Strong learning ability, passionate about algorithms and mathematics, with strong skills in understanding, deriving, and implementation.
Possess solid programming skills, capable of independently completing algorithm implementation, experiment replication, and research code development.
Research or project experience in robotics, machine learning, computer vision, control, reinforcement learning, or related fields.
Possess certain experience in mathematical modeling, have participated in mathematical modeling competitions, research training, or have modeling practice of equivalent intensity.
Familiar with modern AI-assisted development workflows, with hands-on experience using tools such as Codex, Claude Code, and Gemini CLI, capable of improving experimental, development, and research efficiency.
Familiarity with VLA, OpenVLA, and other related models or frameworks is a basic requirement, not an extra credit; we hope you can read the relevant papers, understand the basic paradigms, and have a certain level of hands-on and experimental ability.
Have a basic understanding of robot body, control constraints, sensory inputs, and real-world deployment issues, and understand that robot algorithms are not just purely offline modeling.
Possesses a good sense of scientific research collaboration and is willing to engage in literature reading, experiment replication, result analysis, and continuous iteration.

Has experience in publishing papers, writing technical reports, or producing relatively complete scientific research outputs.
Participated in robotics competitions such as RoboMaster (RM) and has undertaken work related to algorithms, perception, control, or system integration.
Has experimental experience on a real robot platform, not just simulation, offline data, or pure model training experience.
In addition to VLA and OpenVLA, there is further practice with related models, frameworks, or data paradigms such as ACT, LeRobot, OpenPI, and GR00T.
Has conducted relatively complete research projects, model reproductions, or systematic experiments, and can clearly explain their own contributions and conclusions.

Submit your application below and our team will review your resume shortly.